基于机器视觉的激光分层除漆清洁度在线监测方法  

On-Line Monitoring Method of Laser Delamination Paint Removal Cleanliness Based on Machine Vision

作　　者：张赛 杨文锋[1] 林德惠 曹宇[2] 胡月 张然 李绍龙 Zhang Sai;Yang Wenfeng;Lin Dehui;Cao Yu;Hu Yue;Zhang Ran;Li Shaolong(Civil Aircra ft Composites Maintenance Research Center,Civil Aviation Flight University of China,Guanghan 618307,Sichuan,China;College of Mechanical and Electrical Engineering,Wenzhou University,Wenzhou 325035,Zhejiang,China;The Second Research Institute of Civil Aviation Administration of China Chengdu 610000,Sichuan,China)

机构地区：[1]中国民用航空飞行学院民机复合材料维修研究中心,四川广汉618307 [2]温州大学机电工程学院,浙江温州325035 [3]中国民用航空局第二研究所,四川成都610000

出　　处：《应用激光》2025年第1期175-183,共9页Applied Laser

基　　金：四川省科技厅重点研发项目(2021YFSY0025);四川省通用航空器维修工程技术研究中心项目(GAMRC2021YB07);大学生创新创业训练计划项目(S202210624055)。

摘　　要：针对飞机蒙皮局部或特定区域激光分层除漆效果的在线监测问题,提出基于机器视觉的激光分层除漆清洁度在线监测方法。基于搭建的激光除漆视觉在线监测平台,获取激光清除面漆过程的图像。在Canny、膨胀、透视变换等算法提取局部或特定激光分层除漆区域的基础上,将提取的激光分层除漆图像从RGB空间转换到HSV空间,之后采用OTSU算法分割V通道图像不同颜色漆层,进一步利用二值算法计算不同颜色漆层面积占比,最终实现激光分层除漆清洁度的监测并验证监测算法的准确性。结果表明,机器视觉监测的激光分层除漆区域清洁度为39.80%,设计实验验证该监测方法的准确性,监测的验证图像面积占比理论值与监测值误差为0.42%,吻合度较好。所提出的方法可实现激光分层除漆清洁度的准确监测,为飞机蒙皮激光除漆在线监测提供方法指导与理论支撑。To address the issue of online monitoring of the effectiveness of laser delayering paint removal in localized or specific areas of aircraft skin,a machine vision-based method for online monitoring of the cleanliness of laser delayering paint removal is proposed.Using an established visual online monitoring platform,images of the laser paint removal process are acquired.First,algorithms such as Canny edge detection,dilation,and perspective transformation are applied to extract localized or specific areas of laser delayering paint removal.Next,the extracted images are converted from the RGB color space to the HSV color space.The OTSU algorithm is then used to segment the V channel image into different color paint layers.Subsequently,the ratio of the areas occupied by these layers is calculated using a binarization algorithm.Finally,the cleanliness of laser delayering paint removal is monitored,and the accuracy of the algorithm is verified.The results show that the machine vision-monitored cleanliness is 39.80%.Experiments demonstrate a 0.42%error between the theoretical and monitored area ratios,indicating high consistency.This method provides methodological guidance and theoretical support for online monitoring of laser paint removal on aircraft skin.

关 键 词：飞机蒙皮 激光除漆 色彩转换 清洁度 在线监测 

分 类 号：TN249[电子电信—物理电子学]